The Thrill of Discovering Patterns
We love patterns and repetition. They are pleasurable to experience when they are obvious, and equally enjoyable when they are at first hidden. When we perceive a pattern, the rush of pleasure we feel has been likened to orgasm by psychologist Alison Gopnik.1 Patterns across space, such as the textures of walls, are important because they often designate contiguous surfaces in the world. Camouflage works because it disrupts the normal visual pattern associated with the surfaces it is on.
Patterns in time, as we experience them in music and in cause-and-effect relationships, are also important. When we notice that one thing follows another with regularity, we learn to make accurate predictions, which is important for staying alive, among other human goals. In the absence of a teacher or some punishment or reward from our environment, we engage in what artificial-intelligence scientists call “unsupervised learning,” which is, basically, learning patterns in the environment without any immediate feedback. Much of the time, these predictions are correct. I’m talking about very mundane predictions, such as predicting that your foot will not fall through the floor when you take a step in your bedroom or that the cloud you think you see in the sky really is a cloud. Anytime we recognize anything we are detecting patterns, be it the sound of a dog panting, the way your friend’s smile looks, or the taste of milk.
We evolved to be hypersensitive pattern detectors because the cost of missing a pattern (such as a tiger) is greater than the cost of seeing one that isn’t there. Skeptic magazine editor Michael Shermer calls this “patternicity.”2 Our hypersensitivity manifests itself in a host of psychological biases with names such as the clustering illusion (seeing streaks or clusters in random sequences), pareidolia or apophenia (seeing meaning in random or vague stimuli, such as Jesus’s face in toast), and illusory correlation (seeing a relationship that one expects even when no such relationship exists). Fear of autism contributes to people’s belief that vaccines cause autism, but the other major factor at work is illusory correlation. Children with autism tend to be diagnosed at about the same age, coincidentally, that they get a lot of their vaccines. First they get vaccines and then they “get” autism. It’s a great example of seeing causation where there is only correlation.
I heard a story once of instructor Peter Révész, who teaches the concept of randomness in a brilliant way. He would ask one half of the students to flip a coin 100 times and record the result. He asked the other half to come up with a sequence of heads and tails, 100 long, that looks random. He told the groups to write the two strings of heads and tails on the board and that he would come in and guess which one was actually random. With one look he could tell which is the true random sequence—it’s the one with long strings of heads or tails.3
People underestimate the frequency of long strings of recognizable patterns that inevitably show up in long, random patterns. As a result, when these patterns show up, people think that the underlying causal mechanism that produced the pattern must not be random at all. Then they start making guesses about what that nonrandom process must be.
As we become familiar with a subject, such as a school of painting, or a language, or a musical style, we notice more and more of the patterns that make it up. We can look at this as building a vocabulary of these patterns. When one first hears a new language, one understands nothing and hears only an uninterrupted series of sounds. The sounds might sound different in some ways, but they are in an important sense indistinguishable, because we don’t notice what is importantly different about them. We can’t even tell when one word ends and the next begins. Many times, when it starts raining, there is an arrangement of raindrops visible on the window. Each time it rains the arrangement is different, but they still all look the same (that is, like all the other arrangements of raindrops) because we can’t see any patterns in them. There are no patterns there to detect, and our mind becomes blind to them. When observing people, sometimes one will have the impression that members of other races “all look the same,” when, of course, all racial groups have a great deal of internal variety. Indeed, people have been shown to be more accurate at distinguishing members of their own ethnic group.4
But language does have patterns. After many exposures, the sounds of that language, as well as combinations of those sounds, become recognizable. One sentence will sound less and less like all of the other sentences. Before I studied Chinese, I could not tell the difference between Chinese and Japanese. Now I can tell the difference between Mandarin and Cantonese.
Similarly, music might “all sound the same” until we acquire the musical vocabulary to make sense of a new genre. My grandfather once concluded that kids must like contemporary pop music because of its lyrics, on account of the music itself sounds so similar from song to song. Although it seemed incredible to me at the time that he could think that Duran Duran sounded just like Beastie Boys, to him they did. Of course, even fans of popular music often can’t tell the differences between different genres of electronic dance music until they get into it a bit. Can you tell the difference between house and trance? How about jungle and dubstep? The differences are obvious to fans.
Patterns also allow for efficient memory storage. For example, memorizing a list of letters, such as “klmrdfkfj,” can be challenging. There are nine things you need to remember. However, it gets much easier when those same letters are rearranged as “JFK FDR MLK.” If a person knows which famous people those initials refer to, she needs only remember three pieces of information rather than nine. It makes sense that we would have evolved to be happy about discovering things that are easier to remember, because they require less effort. The storage of perceptions of recognizable patterns makes for a kind of data compression in memory. When a pattern is noticed, only a symbol for the pattern need be stored, rather than the many bits that constitute it. In Standard American English, for example, we often pronounce the first syllable of police differently from the first of Pocahontas. But we hear and notice them as being the same. Similarly, when we look at writing, we learn which marks are extraneous and which are important, allowing us to read the same words in different fonts. People unfamiliar with our alphabet have trouble doing this, because they have not yet learned the essences of the characters.
Perhaps the pleasure we feel when experiencing patterns is a part of what we call beauty. Artificial intelligence researcher Jürgen Schmidhuber’s complexity-based theory of beauty holds that things are more beautiful if they require less memory storage because they are simpler.5 Experts see more patterns than nonexperts. As you might expect, experts are more easily bored with simple stimuli and prefer more originality. Art connoisseurs prefer more abstract and conceptual paintings.6 Car experts prefer cars with unusual designs. This effect has even been replicated in laboratory experiments by psychologists Claus-Christian Carbon and Helmut Leder—simply exposing people to innovative car designs again and again makes them, in a sense “experts,” who then prefer the more innovative designs.7
Symmetry, particularly left-right or bilateral symmetry, is compelling because of pattern, and is in fact the strongest predictor of beauty judgments in basic shapes and patterns. Our preference for bilateral symmetry might be explained by the fact that in the natural world, it tends to indicate the presence of a living thing, and gravity dictates that earthbound living things will not have an up-down symmetry, because one side would rest on the ground and another not. Also, when an animal appears as bilaterally symmetric, it is more likely to be facing toward or away from you, because a side view of an animal is not symmetrical.
The detection of symmetry is the detection that something is the same on both sides or all around. It requires ignoring minor differences spread across space. Dyslexia is a disorder characterized by mixing up letters that are the reverse of one another, such as the lowercase b and d. Dyslexics see these letters as symmetrical and therefore more similar. This doesn’t occur to others, and indeed a study by psychologists Thomas Lachmann and Cees van Leeuwen found that dyslexics enjoy the benefit of having better symmetry detection and are also more likely to be able to read upside-down.8
Developmentally, it is difficult to learn to make right-left distinctions. We all start out as kind of dyslexic. In the environment of our evolutionary adaptation, this distinction wasn’t particularly important. We have to partially unlearn symmetry detection in order to read properly (to distinguish, for example, a lowercase b from a lowercase d).9 This suggests that dyslexics would find symmetrical art even more compelling than everyone else.
One incredible finding in psychology is that we are more likely to like and even believe things that we find easy to understand. Before explaining why this is true, I’ll go over some of the evidence that this perceptual fluency hypothesis actually is true.
Things that are easy to understand require less mental processing, which affects the way stimuli are judged. If there is less cognitive processing involved, the idea or experience in question is easier on the mind. Stimuli that are easier to process are perceived as more familiar and visually clearer, more pleasant, louder, longer, more recent, more likely to be chosen, less risky, more attractive, and more truthful.10
Statements made in an unfamiliar accent are more difficult to understand and, as we’d predict, people actually believe them less, as shown in a study by psychologists Shiri Lev-Ari and Boaz Keysar.11 Difficult-to-read text causes a kind of mental speed bump that makes us stop and think more critically, according to an experiment by Jesse Lee Preston. It showed that when reading arguments in favor of capital punishment, people exhibited less of a right- or left-wing bias when the font was difficult to read. People reading about cases in a mock trial were more evenhanded if the text describing the crime had been visually degraded.12
People like simple explanations and simple solutions to problems. If ideas are too complicated, people have a tendency to reject them. Psychologist Dean Keith Simonton’s study of American presidents and British prime ministers shows that to get elected there is a critical window of intelligence into which the candidate must fall. They can’t be too dumb or too smart, and where this window lies is based on the intelligence of the voting population. In America, everyone can vote, so the ideal IQ of presidents tends to be around 119. In the United Kingdom, the House of Commons elects its top leader. Because members of parliament tend to have a higher IQ than the general populace, the window is in a different place. The result is that Britain’s prime ministers tend to have higher IQs than American presidents—a 15-point difference.13
What makes something easy to process? One predictor is familiarity. The more often we see something, the more familiar it is, and the easier it is to process when we see it in the future. Recall that when we are able to detect patterns, we can efficiently store them into memory. This effect begins even after a single exposure to a particular stimulus—simply having experienced something makes it preferable and more attractive. Repeated exposure to something makes it more prototypical. Prototypical images are the ones we associate most strongly with a given category or word. For example, an equilateral triangle is a more prototypical example of a triangle than one that’s two inches wide and a mile high. Predictably, prototypes are considered more attractive. Specifically, people classify prototypical patterns faster and recruit fewer mental resources to perceive them.
Exposure and familiarity even has an effect on what we believe. Experiments by psychologist Ian Maynard Begg show that when people see sentences again and again, they are more likely to rate them as true—even if, when they initially heard the sentence, they were told that it was false.14
Of course, there is a limit to how much exposure can increase positive feelings. Eventually we get bored of the same ideas and experiences. We get fond of things, then we get tired of them, and eventually we might want to return to those things.
The proximate reason for the exposure effect seems to be that being exposed to things makes them more fluently processed. But what is the ultimate explanation? It seems that merely experiencing something that we recognize comes with an automatic positive association.
Being able to detect patterns in the world is absolutely crucial for survival. Pattern detection is required for classification, for understanding causal effects, for learning how to act in particular situations. It’s no wonder we feel a surge of pleasure when we see a pattern—our minds have evolved to reward us for discovering regularities in the world that might be exploited.
* * *
We evolved to find patterns in the world to help us make predictions, and arts of various kinds all take advantage of this, fooling the mind into thinking it’s discovered something about the world, an experience we find pleasing. The easiest example to understand is, perhaps, color. Seeing the same color over and over again in a visual stimulus makes it look better, be it in a website or a match between a woman’s eyeliner and blouse color.
Our writing systems are the way they are because the letters are visually distinguishable from each other and because they look natural to us. In fact, the topological properties of writing systems match those of visual scenes from the world. It could be that rather than our brains evolving for writing systems, our writing systems evolved to fit the brain, as a study by neuroscientist Mark Changizi suggests.15 As the shapes we see in letters are common in the natural world, it’s easier for us to process them. As a result, they are pleasing to look at. As evidenced in the popularity of calligraphy, in many cultures, across many eras, we find letter forms beautiful.
* * *
A possible reason narratives are compelling is that they make sense according to the structures we use to understand the world. Take, for example, the sentence “The king died and the queen died.” Compare that to “The king died and then the queen died of grief.”16 Although the second one is more complicated, in that it has more words and information, in another way it is less complicated. The events in the latter sentence are not disconnected but fit together in a way we can understand. Because it fits into our notions of how people behave, we can almost predict the second event from the first, making the whole thing easier to remember. Of course, another reason the second sentence is more memorable is because it contains social causes.
Writers recognize that repeating symbols draw the reader in. In the world of theatrical improvisation, satisfying endings often “reincorporate” elements from the beginning of a scene.17
Foreshadowing is a way of using repetition to interest the audience. Sometimes foreshadowing is obvious. In film this is called a “leave-behind.” Literature sometimes refers to it as “Chekhov’s gun,” because the playwright Anton Chekhov wrote that an artist “must not put a loaded rifle on the stage if no one is thinking of firing it.” Symbolism also can take advantage of repetition. When we first encounter the paperweight in George Orwell’s 1984, we might not pay attention to that. But when the Thought Police arrest Winston Smith, it shatters on the floor. The power of the symbol is evoked by the memory of the role of the paperweight earlier in the novel.18
We can also find some things people say very compelling. People talk all day, but only some utterances are repeated and turned into quotations and idioms, while others fade from our cultural memory. Why?
Certain quotations and idioms are sticky; they have a ring to them that catches in our memories or appear to say things that strike us as true. Aesthetic principles that govern the appreciation of other arts also apply to quotations. We love symmetry (e.g., “love the way you want to be loved”); incongruity (“keep your friends close and your enemies closer” or “the bigger they are, the harder they fall”); and repetition, be it in rhyme (“see you later, alligator”), alliteration (“a dime a dozen”), or word structure (“a penny saved is a penny earned”).
A shocking number of famous quotations were never even said. What happens is that somebody, usually somebody already famous, says something and the quotation gets altered to be more memorable, more succinct, more compelling. For example, Albert Einstein is famous for having said of quantum mechanics “God does not play dice.” Actually, what he said was the much more cumbersome “It seems hard to sneak a look at God’s cards. But that he plays dice and uses ‘telepathic’ methods . . . is something that I cannot believe for a single moment.” Doesn’t have the same ring, does it?
Titles of books are much shorter than they used to be. Short book titles correlate with the growth of the publishing industry, perhaps because short titles are easier to market in a crowded marketplace. This was found based on a study of 7,000 books in eighteenth- and nineteenth-century England by literary scholar Franco Moretti.19 Short titles are easier to digest, and long titles are so unusual now that they sound humorous (e.g., A Voyage to the South Sea, Undertaken by Command of His Majesty, For the Purpose of Conveying the Bread-Fruit Tree to the West Indies, In His Majesty’s Ship the Bounty, Commanded by Lieutenant William Bligh was the title of a book describing the mutiny on the Bounty). The actual title of what most people consider The Origin of Species is On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life. Just as we make quotations shorter, we seem to do it with titles too. Why? Because ease of processing often increases preference and shorter titles are easier to process.
The value of repeated elements is strikingly obvious in music. Music scholar David Huron conducted a study finding that 94 percent of musical passages are repeated elsewhere in the work.20 Musical repetition works on at least six different levels of abstraction. Perhaps the simplest pattern is the use of the same instruments throughout a piece.
A rhythm can’t even exist without repetition. Beats have time between them, and a constant beat is a repetition of the amount of time between beats. Even a gradually accelerating rhythm has its own kind of pattern, as the times grow linearly or otherwise.
At another level are musical motifs, which are repeated elements, usually melodic. In his operas, Richard Wagner used a motif for a particular character again and again. This technique is used to great effect today in film scoring.
Music with lyrics employs the same kinds of elements as described above with poetry. In addition, lyrics match the rhythm in music. In fact people subconsciously choose words that rhythmically fit with what they are saying even in normal speech.21
Verses and choruses also repeat, allowing us to become familiar with them over the course of a piece of music. And of course we all have had the experience of appreciating a piece of music more and more as we hear it.
* * *
Music creates patterns over time to capture our attention. But even little bits of text use repetition to stick in our memory. As in poetry and song lyrics, compelling idioms and quotations often make use of rhyme, such as “birds of a feather flock together.”
They don’t just sound better; people actually find idioms that rhyme more believable. For example, a study by psychologists Matthew McGlone and Jessica Tofighbakhsh found that people will judge “woes unite foes” to be a more accurate description than “woes unite enemies” because the rhyme increases processing fluency.22 Political slogans, too, sometimes use the same phonetic tricks that poets use, as in the wildly successful “I like Ike.” This has cleverly been called the “rhyme is reason effect.” Why is it clever? Because we’re probably more likely to believe in the effect when the effect has such a great name. It has a pattern in that it reminds us of the saying “no rhyme or reason” and the two content words start with an r (that is, are) sound.23 But it’s a little bit incongruous in that it differs slightly from “rhyme or reason,” piquing our interest.
* * *
Repetition also helps explain our love of sports. If we look at the rules, we can see that they define a structure that includes a good deal of repetition. In hockey, a puck is dropped between two players. In soccer, when a ball goes out of bounds players throw it back in with their hands. In basketball, there are foul shots. These events occur over and over in games. These repeated behaviors are familiar to spectators, who simultaneously enjoy the ritual and rejoice in the nuances of how they play out each time.
* * *
Because finding patterns in the world is so important, we evolved to experience finding a pattern pleasurable. This is the ultimate explanation. But what is the proximate explanation? The brain uses a chemical called dopamine to recognize patterns. Dopamine is a neurotransmitter, one of the chemicals that the neurons in your head use to communicate with each other. Not only does it matter which neurons are passing information, but it also matters how they transmit it—that is, which neurotransmitters actually carry the message. Pattern recognition is one of the many functions of this fascinating chemical.
When the dopamine pattern-recognition system is working, you get high-quality beliefs out of it. However, we’ve all had the experience of seeing a pattern that is not there—a figure in the shadows or a face in the clouds. Dopamine is the neurotransmitter that tags perceptions as meaningful, prompting us to make sense of them.
It is likely that the sensitivity of our pattern detector is in part determined by our dopamine levels. High dopamine makes everything look significant. People suffering from schizophrenia overproduce dopamine, and see patterns even in random noise and notice even the loosest of connections between ideas.24 Research psychiatrist Shitij Kapur theorizes that when we think something is important, our brain struggles to rationalize an explanation for it. He calls this “biased inductive logic.” We try to figure out why the stimuli feel so important. As a result, the explanations are culture specific (it’s the doing of the FBI, a curse from a witch doctor, etc.).25 This is profound. It’s not that a schizophrenic patient reasons that the FBI is controlling his or her mind and then the feeling follows. The feeling comes first, unconsciously, feeding up from the dopaminergic system. Only then does the rational, deliberate part of the mind concoct some story to explain the feeling.
But you don’t have to have schizophrenia to see patterns where none exist. People who naturally have more dopamine are better at seeing faces in jumbled images and are also more prone to see faces that are not there.
People who believe in the paranormal tend to be bad judges of randomness. Suppose you roll a die three times, and you get 5, 1, and then 3. Suppose someone else rolls a die three times and gets 2, 2, and 2. Both results (in those specific orders) are equally probable, because the dice rolls are independent. However, a study by psychologist Peter Brugger found that believers in the paranormal are more likely than skeptics to think that patterns like 2–2–2 are less likely to occur.26
Does dopamine cause the pattern finding? In another study Brugger found that skeptical people tend to have lower dopamine levels and those who believe in the paranormal have higher dopamine levels. In fact, artificially increasing dopamine in skeptics (in pill form) starts making them see more patterns, including nonexistent ones. The dopamine level thus causes the effect, and not vice versa.27
Superstitious learning is an effect discovered by behaviorists in psychology. You can train rats and pigeons to press a bar to get food. This is called “reinforcement.” However, if you give them food at random intervals, a very interesting thing happens: the animal will start displaying odd, idiosyncratic behaviors to get the food. Here’s how it works: suppose a hungry rat happens to be scratching its right ear just before the food arrives. The rat’s mind is desperately trying to figure out some pattern to the food arrival. It assumes (unconsciously, probably) that the scratching led to the arrival. So it keeps scratching. Such animals are acting as though they believe that some combination of what they did resulted in food delivery. They are inaccurately inferring cause-and-effect relationships. If you test ten rats, you’ll get them all doing different superstitious behaviors, based on whatever behavior they happened to have been doing before the food came.
Normally, the pattern-detection mechanism in our minds is anchored in reality—that is, we tend to perceive patterns that are real, in whatever sense patterns can be considered real. But especially under conditions where the link with reality is compromised—such as when the subject is high on drugs, dreaming, under high stress, or under sensory deprivation—the system starts making wacky conclusions about the world and seeing random events as meaningful and related to each other. Deny the pattern detector its normal input and it starts trying to make sense of essentially random or missing inputs: a person might stare at a television on channel fuzz and try to make out a picture, or they might make mountains out of molehills in the dark, misinterpreting the random firing of the sensors in their eyes.
We all differ in the number of hippocampal neurons we have, and that number might be related to our ability to see patterns. Perhaps both our dopamine levels and our hippocampus sizes predict pattern detection and religiosity.
Based on this, I conjecture that religious and paranormal believers are more likely to get schizophrenia, which is characterized by an excess of dopamine, and skeptics are more likely to get Parkinson’s disease, which is characterized by a shortage of dopamine.
Schizophrenia has a genetic component. A milder expression of these genes can result in a condition called “schizotypy,” which is characterized by social withdrawal, odd behavior and thinking, magical thinking, and occasional quasi-psychotic episodes with hallucinations and delusionlike ideas. The hallucinations are milder—more like strange perceptual experiences, such as seeing spirits in a meadow.
Why hasn’t evolution removed these genes from the population? The replication of the gene for schizophrenia-spectrum disorders might be due to the fact that in many traditional societies, schizotypals or epileptics are often the witch doctors and shamans, or they are perceived as blessed and are more likely to set the religious tone of the society.28 Schizotypy is more common in religious people and cult members than in the general population, although it bears noting that some research finds psychotics in general tend to be less religious (schizophrenia is a kind of psychosis). Research by psychologists Rachel Pechey and Peter Halligan shows that people who experience delusions tend to have more paranormal beliefs.29 As anthropologist Alfred Kroeber says: “The shaman displays his possession by a spirit by publicly re-enacting his specific personal experience, that of a man suffering from a particular mental affliction. His projections, his hallucinations, his journey through space and time, thus became a dramatic ritual and served as the prototype for all future concepts for the religious road to perfection.”30 Schizotypals also tend to have very concrete religious beliefs. Most shamans are not celibate, which allows their genes to spread. This suggests not only that schizophrenia-spectrum effects might account for some of people’s continuing beliefs in religion, but that schizotypals might even have originated many of the beliefs religious people hold in culture today.31 The relatives of schizophrenics tend to be more creative and more religious, replicating the genes in the population without displaying full-blown schizophrenia.32
Schizotypalism affects about 3 percent of the population,33 but the problem of perceiving nonexistent patterns is so widespread that it goes by several names in different disciplines. In statistics, claiming that you’ve found a pattern when there isn’t one is called a type 1 error. The literature on psychological biases describes the finding of clusters or streaks in random data, such as basketball shots and coin tosses, which is called the “clustering illusion” or “illusory correlation.” A related effect is subjective validation, a bias that makes someone more likely to see a relationship between two events because such a relationship would have personal meaning to the perceiver. Just because different scholars have come up with different terms does not mean that they are all separate effects. It could be that all these observed effects are caused by the same underlying psychological and brain mechanism. If this is true, then dopamine levels and hippocampus size (as caused by many factors) are good candidates for what this underlying mechanism might be.
Could beliefs in the paranormal and many religious beliefs be linked to an overly amped-up sensitivity for pattern detection? It is likely. In addition to the evidence above about schizotypals, psychologist Paul Rogers found that people who believe in the supernatural are also more likely to be susceptible to the conjunction fallacy (believing, mistakenly, that conjunctions of facts are more probable than the individual facts themselves, such as believing that someone is more likely to be a Republican banker than a banker), to have poor reasoning with probabilities, and to have more misperceptions of randomness.34 That is, people who believe in such things also have reasoning problems in their perceptions and interpretations of other, more mundane aspects of the world.
Skeptics are more likely to miss patterns that really are there, where believers might see too many patterns.
Problems with overactive pattern detection can affect religion, but pattern helps religion maintain itself in a more obvious way—by repetition of rituals and communications of the core belief systems. Many religions require weekly or even daily readings from a sacred text or attendance at clergy sermons. The themes and beliefs of these religions are heard over and over by the congregations. This repetition has obvious implications for ease of processing and memory, ultimately facilitating belief.
People tend to look for explanations for the unusual or unexpected, particularly misfortune. As noted earlier, religions the world over explain misfortune as being the result of punishment or retribution from gods or witches.35 This is particularly likely if several misfortunes occur at once. Because we are bad at detecting randomness, and think we see patterns that are not there, our minds crave an explanation.
* * *
There is good reason to think that psychological factors influence the belief (and disbelief) in conspiracy theories. A study by Viren Swami found that people who believe in one conspiracy theory tend to believe in others.36 This suggests that there can be something about you, some trait, that makes you prone to conspiracy theorizing. Or, for nonbelievers, prone to skepticism.
People of lower social status should be more prone to conspiracy theories for two reasons: first, they are probably less privy in their day-to-day lives to knowledge of how power actually works in the world. This not only leaves a gap of knowledge that the mind wants to fill but also makes the workings of society appear chaotic, which encourages superstitious reasoning. And recall that people primed to feel out of control are more likely to see patterns in random stimuli.
Secret knowledge of how the world works would also be more valuable, as such people have further to go up the social ladder than a higher-status person. Indeed, people who endorse conspiracy theories are especially likely to feel angry, mistrustful, alienated from society, and as if larger forces are controlling their lives. They feel out of control.
Conspiracy theorists are also often immune to any kind of counterevidence. As we’ve seen, when they encounter contradictory evidence, they don’t see it as evidence that disproves their theory, but as further evidence of the cover-up. People are creative and can figure out how to fit new information into their conspiracy theory narratives. Ironically, the smarter a person is, the better they are at it! For example, a conspiracy theorist might believe that aliens have visited earth and that the United States government is covering it up. If the government releases information explaining that a particular UFO sighting was a secret aircraft owned by the military, conspiracy theorists will classify that press release as a cover-up. As people classify things according to their theories, it feels to them that the theories are gaining evidence (religious people also feel they have experienced a great deal of evidence for God’s existence as a result of classifying so many day-to-day things as being divine interventions). When evidence and counterevidence both count as evidence for the theorist, it’s easy to see how everything they experience reinforces the explanations they already have. Talking to a conspiracy theorist can be like playing table tennis in a room with an oscillating fan. A bit of advice to the reader: when debating with a conspiracy theorist, don’t debate evidence; they will have an answer for everything. Debate their epistemology—try to show them that they have set up a way of thinking that is immune to any conceivable counterevidence. They are trapped in a vicious circle that’s very hard to escape.
As evidence against the conspiracy theory grows, the theory itself becomes more complex to account for the growing reasons to disbelieve it. Conspiracy theories are complex and constantly changing. When scientific theories develop ad hoc additions to account for any state of the world, it is seen as a problem. The theory loses parsimony and it loses ground to other, simpler theories that explain the same observations. Unfortunately, this does not seem to happen with conspiracy theories.
As the complexity of the conspiracy grows, theorists are ever more fascinated. They can feel this fascination even if they don’t believe in the theory. For many, the wild and unlikely theory that the government constructed HIV in order to kill gays and blacks is far more fascinating than how HIV actually became a problem. The bigger and more convoluted the conspiracy theory is, the more important it feels.
A mistrust of authority, what Michael Shermer calls the “heretic personality,” is a hallmark of the conspiracy theorist.37 So deep is this mistrust that many are willing to believe contradictory views, as long as both of those views are different from what the authorities are telling the people. An interesting study by psychologist Michael Wood showed that if someone believes that Princess Diana faked her death, they are more likely to also believe that she was murdered.38 If people are willing to entertain contradictions in their own theories, how can counterevidence have any chance at all of changing their minds when it’s presented?
This is one area in which religious people and conspiracy theorists differ—the religious tend to trust authority more than the average person.39 In one experiment by psychologist Daniel Wisneski the more religious participants tended to trust the ability of the Supreme Court to make good decisions whereas participants with strong moral, but not religious, convictions felt distrust.
* * *
Ideas that seem absurd at first become more plausible after repeated exposure. When an idea gets reported repeatedly in the media, or at church, or by the people you socialize with, it has an effect of making the idea seem more believable. Why does the “availability cascade” happen? I suspect there are two processes at work here. The first is the “bandwagon effect,” which makes people do and believe things simply because other people are believing and doing them. The reason most relevant to this chapter, of course, is that repeated exposure creates a pattern, increasing cognitive fluency, resulting in a favorable evaluation. We simply like the idea more. When a person repeatedly hears about an idea, there is social pressure to accept the idea (we don’t want to challenge our peers) and there is a cognitive pressure to accept the idea (common ideas are just easier on the mind). What might appear unusual and weird at first becomes familiar, softening our minds to the idea.
The availability cascade can lead to good or bad outcomes, depending on the quality of the idea. As an example of a bad outcome, let’s look at the frequency of child molestation in day care centers and compare it to the terror felt by parents. During the 1980s and early 1990s, people started to be very frightened that their children would be sexually assaulted in day care centers. Some well-publicized examples made the problem seem more common than it really was—a study at the time found that kids were actually in more danger in their own homes than at day cares.40 This fear still lingers today.
One study in the United Kingdom found that just 2 percent of caregivers for children under five are men.41 An anticipated fear that male childcare workers are likely to be child molesters has likely had a crippling effect on men’s abilities to get jobs as childcare workers. Is there anything wrong with this? There might be. Oxytocin is a hormone that causes bonding between people. When mothers cuddle their children, they get a boost of oxytocin. Men, relative to women, get the boost when engaging in more rough-and-tumble play. The theory is that men and women evolved to provide the offspring with different experiences. With only women in day care centers, society seems to have inadvertently made the decision that we don’t need men to help raise children outside of the home. It is time we asked ourselves if this is really what we want.
As conspiracy theories and ideas of alien abduction get more popular, they become accessible narratives in people’s minds. Then, when people have new experiences or hear new stories, those narratives are there for people to potentially fit the new information into.
Let’s look at the common view of alien abduction. The idea that aliens are visiting our world and abducting people has been raised over and over in Western culture. When you hear about alien abduction there’s a part of your mind that starts to think there’s something to all the hype. In the media, as in rumor, this can be a self-strengthening cycle. When one seeks out a group of other believers, for example, proclaimed abductees, one becomes part of a subculture where the availability cascade becomes even more intense, as the ideas get recycled with more regularity and less counterpoint.
* * *
It’s obvious how the availability cascade works with religion. People who grow up in a culture where nobody questions the existence of gods and spirits rarely question their existence. This is true for religious beliefs, as well as for interpreting events (e.g., pregnancies, deaths, temporal lobe seizures) as having religious significance.
One morning I was lying in bed and out of the corner of my eye I perceived a shadowy thing moving by my window. I looked over and saw nothing. I don’t believe that there are spirits that can manifest themselves in this world. I do believe that the peripheral areas of the eye are sensitive to movement and that perception is sometimes subject to random fluctuations that can result in misperceptions—which is exactly how I interpreted the situation. My “perception” of something moving was, in my interpretation, a glitch in my visual system, not a spirit.
However, people who believe that we are surrounded by spirits we can sometimes see might have interpreted this experience very differently. Given their beliefs, it would be rational to assume that this could very well have been the experience of a spirit.
I met a scientist once who believed in magic. When I asked him why, he shook his head at me, incredulous, and said, “Just open your eyes.” He saw magic happening every day. Believers and nonbelievers alike have open eyes. What’s different is how they understand what they see. We interpret our experiences in a way that makes sense given our beliefs about the world. He and I both remember our experiences as being further evidence of the very belief systems that generated them. The problem, in both cases, is that we never got any outside confirmation that our interpretation was correct. The result is that two people with the same perceptual input end up remembering them as confirmations of completely different world views. It’s no wonder it’s so hard to change people’s minds, skeptics and believers alike.
Over the years, the “understanding” becomes unconscious. It feels like we are perceiving something directly. For example, in a particular culture someone might fall into a trance. The people who witness this event feel that they are witnessing a possession by a spirit. To them, it’s as obvious as when we see someone eating enthusiastically and “see” that someone is hungry, or “see” that a child really likes his ice cream. These are assumptions at an unconscious level that feel just like regular perceptions.
Often patterns are hard to see, and we don’t perceive them at all unless we’re looking for them. We can deliberately change how sensitive we are to patterns. Cognitive scientist Jesse Bering puts it well:
When people ask God to give them a sign, they’re often at a standstill, a fork in the road, paralyzed in a critical moment of existential ambivalence. In such cases, our ears are pricked, our eyes widened, our thoughts ruminating on a particular problem. God doesn’t tell us the answers directly, of course. There’s no nod to the left, no telling elbow poke in our side or “psst” in our ear. Rather, we envision God, and other entities like him, as encrypting strategic information in an almost infinite array of natural events: the prognostic stopping of a clock at a certain hour and time; the sudden shrieking of a hawk; an embarrassing blemish on our nose appearing on the eve of an important interview; a choice parking spot opening up at a crowded mall just as we pull around; an interesting stranger sitting next to us on a plane. The possibilities are endless. When the emotional climate is just right, there’s hardly a shape or form that “evidence” cannot assume. Our minds make meaning by disambiguating the meaningless.42
In cultures where there are differences of opinion regarding religious matters, there are competing cascades. Public opinion can rapidly change if at least 10 percent of the population actively proselytizes, as a simulation conducted by Jerry Xie showed.43 It is a common cult strategy to forbid members from communicating with anyone outside the cult. This turns the normal availability cascade into a prison for the mind.44 The religion’s leaders might know explicitly that the taboo of talking to those outside the cult will help the cult survive, in that talking to nonbelievers might undermine faith. The members of the religion might also know it implicitly, in that they might not be aware of the function of such rules, but the religion holds them nonetheless because it was an adaptive trait that helped it compete with other religions as it evolved. Many religions can be successful without this strategy, which raises the question of why the taboo works well for some religions but not others.
With the Internet, it’s easy to find people who agree with just about any belief. People who disagree with us can be annoying and there’s a temptation to not talk to them—it doesn’t matter if the subject is religion, politics, child-rearing practices, or whether the Star Wars prequels were any good. But if you decide to stop talking to people who disagree with you, know that it’s hard to ever change your views, even if you should.
Patterns also make religion attractive because of the frequency of patterns of behavior: rituals. The repetition of actions, including recitation of meaningful text, reinforces the familiarity (and thus, the believability) of the religion’s doctrines. In fact, repetition in a ritual increases its perceived effectiveness. Even nonbelievers derive comfort from ritual, perhaps because it gives them a feeling of control, as psychologists Michael Norton and Francesca Gino found.45 Of course, too much repetition leads to boredom. This “tedium effect” has been reported by anthropologist Harvey Whitehouse as a problem for missionaries trying to evangelize.46 As we will see in the next chapter, as alluring as patterns are, some incongruities are necessary to keep us riveted.
One common explanation for why religion exists at all is that it provides explanations for phenomena we don’t understand. Anthropologist Pascal Boyer believes that many religious beliefs are attempts at explaining our own intuitions. Because much of our mind’s processing is hidden from our consciousness, we feel that ideas and feelings just pop into our heads from nowhere.47 This is true for moral intuitions as well as artistic ideas. Such things can be more easily understood if we believe they are coming from some outside source, such as a god or a muse.
When we hear stories, be they religious or otherwise, they can be hard to remember if they contain too many unfamiliar elements. Our minds tend to replace some of them (for example, a canoe) with a more familiar one (a rowboat) as found in a classic study by psychologist Frederic Bartlett.48 Jokes about old presidents get changed to jokes about new presidents. As people revise ideas to better suit their cultures, and what they are familiar with, the ideas are more likely to be remembered and passed on to other people.
* * *
Supernatural agents tend to fall into certain types because they fit patterns we understand well. Compelling supernatural agents align with the institutions we already have of objects, animals, and people, usually with only one or two characteristics from another category. We have core theories of knowledge about the world—physical, social, and perhaps biological, and many compelling beliefs are the result of mixing the intuitions from them, such as applying beliefs from social knowledge to the physical world. Examples of this mixing include ghosts, a tree that can hear you, or a statue that bleeds. Such ideas are more common in religion than, for example, a god who only exists three days out of the week, because that is something that neither tools, animals, people, nor objects tend to do. Studies have shown that the best candidates for religious supernatural agents are those that are clearly in one category, but with only one or maybe two characteristics from a different category. More just muddies the waters. These strange but yet fairly uncomplicated ideas, called “minimally counterintuitive,” are more interesting, memorable, and easy to communicate.49 They’re riveting.
Some ideas are “mundane concepts.” They are so normal sounding as to just be plain believable—and forgettable. If someone were to say “I know a woman named Vanessa with brown hair,” nobody would ever have reason to doubt this. On the other hand, they also would not find this fact particularly memorable or interesting.
Somewhat more compelling are what can be called “merely counterfactual concepts,” which are probably not real, such as “a newborn baby who can cook beans.” This is somewhat arresting, and a better candidate for a religious belief than a mundane concept. But it is not quite as good as a minimally counterintuitive idea because none of the characteristics cross category boundaries. In other words it just happens to be true that babies can’t cook beans, but it’s not that babies are in a general category of things that can’t. Babies are people, and people can cook beans, so it’s not as severe a violation, as, say, a crying statue would be.
The minimally counterintuitive concept of a man who can walk on water is in the sweet spot—it is exactly the kind of idea that makes a religion thrive. It involves enough of what we understand to make it comprehensible, in its own way, but violates what we know to be true in the real world enough to make our minds continue to work on it, to try to figure it out. Like a paradox, it nags at our attention for a long time. We can’t find the closure that would allow us to neatly categorize it, and then forget about it, as we do with many mundane concepts.50
Whereas mundane and counterfactual ideas do not go far enough, “bizarre concepts” go too far. For example, a man who can walk on water, give birth, read peoples’ minds, and create glasses of milk through force of will alone sounds outlandish, and people are less likely to accept that it could be real.51
Likewise, some human beings are thought to be special. Perhaps they can predict the future, or have lived for 200 years. To understand these cases (and I’m talking now about understanding, not particularly believing) someone uses the normal mental template of “person” and changes it. Usually there are only one or two changes. Our “ontology” is our understanding of the basic kinds of things in the world (for example, plants, animals, rocks). A counterontological idea violates what we know and this incongruity draws our attention. Indeed, cross-cultural psychological experiments with novel supernatural concepts support this. Ontological violations were remembered better than mere oddities. A man who can walk through walls is better than a man with six fingers, because the latter does not violate a man’s perceived essence. So although surface appearances can change greatly from culture to culture (this god has horns, that one has three legs) the ontological violations tend to be similar, falling into a few kinds: people or animals that violate physical properties (e.g., incorporeal or invisible), biological properties (e.g., don’t die, result from a virgin birth), or psychological properties (e.g., have supernaturally extended perception). Similarly, objects can have biological properties (such as bleeding) or psychological ones (such as being able to hear). If there is more than one violation, it is more easily forgotten. Multiple violations are rare and tend to be limited to literate theology and are not popular with the public. Psychologists Michael Kelly and Frank Keil studied tales from Ovid and the ancient Greeks and found that mythological and fairy tale metamorphoses tend to happen between kinds of things that are similar.52 For example, in mythology it is more common for people to turn into animals than into plants. And, consistent with the social compellingness hypothesis, the nature of the changed person’s mind tends to remain unaltered. Counterontological beliefs, in small quantities, are more easily remembered, giving them a distinct advantage for cultural survival.
The vast majority of religions have something to say about death. Pascal Boyer believes this comes from counterontological thoughts about dead bodies. When a person dies, especially someone close to us, we have conflicting thoughts. The sight of their body still evokes some of those same ideas of affection, including the person having a mind and our feelings of love for them, yet now our biological intuitions are telling us that the body is an object, and our contagion systems are telling us that the body is dangerous! It is common for religions to hold that a corpse is polluted, or impure. Our minds are torn—it’s a dead animal and at the same time it feels like it’s still a person with a mind. These conflicting thoughts and emotions require some kind of resolution, and religion is up to the task.
Just as we have “object permanence” processes that help us keep track of objects when they leave our field of view, we have “person permanence” processes that allow us to understand that people do not vanish from our world when they walk out of the room. When a loved one dies, our theory of mind and person permanence processes “believe” that the person still has a mind. However, it feels obvious from looking at a dead body that the mind is not there. Religious thinking steps in and generates a belief that the mind has escaped the body, because it feels like the mind is still there. Indeed, we can still feel anger or love directed at dead people. Because the dead body brings up these strange feelings and thoughts, all religions hold that something must be done with the body.
Boyer’s counterontology theory predicts why all religions have things to say about the dead and why spirits of the dead are the most widespread kind of supernatural agents the world over. Cultural inventions such as a statue that cries or a person who is descended from a god are compelling because they cause dissociations with our intuitive systems. Dead bodies are naturally occurring objects that automatically cause counterontology.
After a while, our intuitive minds get used to the idea that the person is actually gone, leaving us only our memories. This is the explanation Boyer provides for “double funerals,” which are very common the world over. In these cases, a body is treated (buried, burned) immediately (for safety reasons, ultimately). Months later the body is unearthed, and the bones are put in their final resting place. For Boyer, these two rituals mark psychological changes in our feelings about the deceased persons’ bodies. The first ritual is for our feeling that the person is not really gone and the second marks the fading of that feeling.
These counterontological thoughts and feelings about bodies explain our fascination and fear with varieties of living and dead people. Boyer says that peoples’ basic concepts relevant to religious agents are life, sentience, and personal identity; in Justin Barrett’s version of the theory people think in terms of psychology, biology, and physics. I suggest that our intuitive notions of people involve four basic components: body, life, animacy, and mind.
My “theory of the living dead” is a speculative description of our fascination with different combinations of these fundamental person concepts. Following Boyer’s counterontology theory, conflicts in intuition should create compelling concepts. You can chart out all the combinations, making 16 rows, of these ontologies. Having a “body” is just what it sounds like—knowledge or perception of a physical body. “Life” is a concept we get from intuitive biology. We experience feelings, at a visceral level, regarding whether a biological object is alive or dead. “Animacy” just means that the body moves on its own volition (as opposed to a rock rolling down a hill). “Mind” means that our theory of mind understands the being to have things like beliefs, goals, perceptions, and feelings.
By looking at every combination of these concepts, we can find compelling supernatural agents from fiction and religion and also combinations that are curiously empty. I will describe a few entries in this grid.
What has life but no body, animacy, or mind? Perhaps some kind of life force. The doctrine of vitalism, now discredited by science, holds that things are alive because they have some kind of life energy. East Asian philosophy posits the existence of qi (Chinese; ki in Japanese). The idea of a life force is a part of natural psychology and many religions.
When a person is sleeping or in a persistent vegetative state, they have a body and life, but not animacy or a mind. Because we are familiar with sleeping, I believe that this particular arrangement is not as compelling or frightening as the others. This observation would predict that people would find a persistent vegetative state less interesting and creepy than either locked-in syndrome or severe mental illness. (Locked-in syndrome is when a person is fully conscious but unable to control their body. These people have a body, life, and a mind, but are missing animacy. A tree that can hear conversations also fits this description.) I cannot think of something that has animacy without any of the other concepts.
The idea of a zombie includes the concepts of a body and animacy, but not life or a mind. This makes zombies appear fundamentally weird, and fascinating. Zombies are popular in speculative fiction, but the basic idea originated in Haitian religion. Sleepwalkers are also interesting, in that they have a body and animacy, but are apparently mindless.
Many other combinations of these characteristics have parallels in folklore, myth, and fiction.
However, just as single ideas cannot be too simple or too complex in order to be acceptable, larger narratives also must fit into a sweet spot. Religious stories as well as fantasy fiction tend to have mostly mundane concepts, with occasional minimally counterintuitive concepts thrown in to keep them interesting. Studies have shown that this is the optimal mix for long-term memory. Too many intuitive ideas or too many counterintuitive ideas result in poorer recall over time. This is true for fiction as well as religion.53
* * *
Ideas pop up, and some of them get around with amazing speed and stick around with surprising tenacity. People are curious and want to understand the world they live in, and many people turn to a belief system to get answers. I turned to science, but many turn to religion and cite getting a better understanding of the events in their lives as a chief reason. Let’s face it: some things in this world are just hard to understand. They’re complicated. Exacerbating this problem is the fact that certain phenomena, particularly at the subatomic or galactic levels, operate according to rules that violate common sense. These two factors make some scientific explanations, frankly, unbelievable.
It’s been said that quantum mechanics is something “our brains just aren’t wired to understand.”54 At the quantum level, objects disappear in one place and reappear in others, have indeterminate locations, and can pop into existence from nothing. It’s hard to believe because it’s just too weird, too unlike the world we live in and understand. We evolved in “middle world” (the perceivable scale of the world, larger than quantum and smaller than galactic), and we’ve evolved to be able to understand how medium-sized objects behave. Indeed, quantum mechanical theory is so strange that physicist Paul Davies said that quantum theories are “almost impossible for the non-scientist to discriminate between the legitimately weird and the outright crackpot.”55
Scientists themselves are not immune to these biases either. In 1979 Carl Woese developed a new classification of life based on data from DNA. It suggested three main branches of life, which he called domains. Animals and plants were just a small branch of one of the domains. Many scientists, especially botanists and zoologists, were slow to accept Woese’s classification. Woese said, “Biology, like physics before it, has moved to a level where the objects of interest and their interactions cannot be perceived through direct observation.”56 We see animals everywhere—doesn’t it make sense that they would be a bigger part of the taxonomy?
Ernst Mayr criticized the new scheme, arguing that one of the reasons to reject the new scheme was “the principle of balance.”57 Balance is a pattern that Mayr found so valuable that it made him prefer one scientific theory over another.
Our love of patterns contributes to our enjoyment of a great variety of experiences. Explanations with patterns, good and bad, attract us for the same reasons. Beware!